Seven years after the identification of the first imprinted genes our understanding of imprinting and its consequences for human phenotypes has improved in some areas but remains incomplete in others. Imprinting has been shown to be a chromosomal domain effect and three imprinted genes H19, IGF2 and p57KIP2, on chromosome 11p15.5, have been shown to be altered by one-hit genetic or epigenetic pathways in Wilms tumor (WT) and Beckwith-Weidemann Syndrome (BWS). Areas of incomplete understanding include the mechanism of domain-wide imprinting, the repertoire of imprinted genes on chromosome 11p15.5, the characteristics of interindividual variations in domain-wide imprinting, and the roles of recently identified and yet-to-be-characterized imprinted genes in growth regulation and tumorigenesis. They will continue to identify new imprinted genes in the chromosome 11p15.5 imprinted domain. The sequences and structures of imprinted genes, including three which they have recently characterized, IPL, IMPT1 and 2G3-8, will be used to test predictions of the host-defense and maternal/paternal competition theories of imprinting. IPL and IMPT1 show intriguing sequence homologies suggesting roles in growth-regulation, so we will carry out functional studies by germline deletions in mice for both genes and, for IPL, biochemical experiments to determine the protein-protein and protein-membrane interactions of its gene product. A subset of imprinted chromosome 11p15.5 genes will be evaluated for participation in the multi-gene WT2" locus. In particular, they will ask whether any other imprinted 11p15.5 genes are affected by epigenetic lesions in WTs analogous to those which affect H19. DNA methylation is important in normal and pathological imprinting of H19, but other epigenetic modifications may be important in mediating domain-wide imprinting. By testing multiple imprinted genes in the domain for susceptibility to imprint erasure by manipulations which alter DNA methylation or histone acetylation we will ask whether the generality of involvement of DNA methylation might be overemphasized by focusing only on the structurally unusual H19 gene as a model system. By measuring allelic expression of genes spaced throughout the domain, we will ask whether strong imprinters and weak imprinters of the entire domain can be identified in human populations; the results will have implications for the mechanism of domain-wide imprinting and, potentially, for variation in human growth phenotypes and cancer susceptibility. Finally, they will test the hypothesis that the untranslated H19 RNA acts in trans to inhibit the translation of growth-promoting mRNAs.